huggingfacetb/smollm2-360m

Provenance & integrity

Revisionsha:f8027fd0…c38b49

Weight fingerprintsha256:7aaff666…1e540f

config.json oidoid:2c111af0…bf16c4

Parameters360M· 361,821,120 params

Licenceapache-2.0

Released2024-11

MakerHugging Face (HuggingFaceTB)

Last modified2025-02-06

Accessopen access

Identity & integrity of the public artifact captured 2025-02-06 — the same file, unchanged, that Ardora read. Not a safety statement.

The reading

Character abstainedthe read did not separate one character clearly at the published ceiling

Disposition

text continuationhigh confidence

plain text-continuer; no assistant register

assistant/chat registerhigh confidence

not detected

Axis profile

Overall · band 1 · high confidence

minimal - a plain continuer with little disposition to characterize (clearly read)

Assistant Adherenceminimal - a plain text-continuer; does not take the assistant turnhigh

Register / Formality— not characterised · the read battery does not calibrate tone / diction; register-formality is not characterized at the published ceiling

Reasoning Scaffoldingminimal - plainly continues text; no staged deliberationhigh

Domain Specializationminimal - general-purpose; no domain-idiom pullmedium

Verbosity— not characterised · default response length / elaboration is not exercised by the read battery; not characterized at the published ceiling

Turn-Taking / Interactivityminimal - one-shot continuation; no interactive turn-takinghigh

Performed Voiceminimal - neutral; holds no character or personamedium

disposition-definition (descriptive): how pronounced and clearly-read the model's overall character is, discounted by what had to be abstained. NOT a quality, capability, or safety ranking. A plain base reads low because it has less disposition to characterize, never because it is worse.

Abstained — read, but not characterised at the published ceiling

domain expertise (medical/legal)insufficient signal to characterize at the published ceiling

multilingual dispositionthe read battery is English-centric; not characterized at the published ceiling

Characterization is descriptive and coverage-bounded; preview engine, single-battery read; see published fidelity ceiling.

Lineage

descendant ← derived_from · sft_dpohuggingfacetb/smollm2-360m-instructInstruction tuning (SFT + DPO) added a clear assistant disposition where the base is a plain continuer. The direction matches the rest of the family.

Claim grounding

The uploader states: "SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters. They are capable of solving a wide range of tasks while being lightweight enough to run on-device." This 360M entry is the base pre-trained model, trained on ~4 trillion tokens (FineWeb-Edu, DCLM, The Stack, plus curated filtered datasets). The card states the models "primarily understand and generate content in English." HF tags: text-generation, transformers, safetensors, en, llama, text-generation-inference (no `conversational` tag).
— the uploader's card, captured 2026-07-07, quoted verbatim · source ↗

SUPPORTEDidentity

This is a base / pretrained language model (the SmolLM2 family base rung), not instruction-tuned.

Ardora reads this as a plain text-continuer with no assistant/chat register (assistant-adherence band 1, high confidence; ardora_score band 1). That is exactly the disposition of an untuned base, so the uploader's "base / pretrained" identity is supported by the reading.

witness wit_smollm2-360m_base · replayable

SUPPORTEDsize

A compact model, lightweight enough to run on-device (the 360M rung of the sweep).

Provenance-corroborated: the parameter count and architecture were verified from the model's config.json and recorded in the atlas provenance. This is a catalogued fact, not a disposition read.

witness provenance:config.json · replayable

UNVERIFIED-ABSTAINcapability

"Capable of solving a wide range of tasks."

"Solving a wide range of tasks" is a task-competence claim. Ardora reads disposition (what the model tends to do), not capability, so it neither confirms nor scores this.

SUPPORTEDdomain

General-purpose; the card names no specialized domain.

The reading finds no pull toward any specialized domain idiom (domain-specialization band 1); consistent with the card's general-purpose framing.

witness wit_smollm2-360m_base · replayable

UNVERIFIED-ABSTAINlanguage

The model "primarily understand[s] and generate[s] content in English."

Ardora's reading is English-centric and explicitly abstains on multilingual disposition; it can neither confirm nor contradict the "primarily English" claim at the published ceiling.

UNVERIFIED-ABSTAINlineage

Pretrained on ~4 trillion tokens (FineWeb-Edu, DCLM, The Stack, plus filtered datasets).

Training-token counts and corpus composition are provenance the atlas records best-effort from the card; Ardora does not independently verify training data, so this is out of the reading's scope.

UNVERIFIED-ABSTAINcapability

Improves over SmolLM1 across knowledge / reasoning / instruction benchmarks (HellaSwag, ARC, MMLU, PIQA...).

Benchmark scores and cross-version capability comparisons are outside Ardora's disposition scope. The reading confirms the base's plain-continuer disposition but does not score capability.

Claims are the uploader's, quoted from their public card at the capture date; stances are coarse, witnessed, disposition-only, measured against the published fidelity ceiling for this engine version -- not the Engine's raw numbers. Recomputed when the ceiling moves; capability, benchmark, and safety claims are abstained, never refuted.